Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses [review]

A recent study comparing body powered and myoelectric prostheses (Engdahl SM, Gonzalez MA, Lee C, Gates DH. Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses. Journal of NeuroEngineering and Rehabilitation. 2024 Aug 8;21(1):138 [1])  provides and interesting glimpse into the academic approach to the subject matter, which taps on a reality, that, from the trenches of real task and work use as I experience and see them, looks different.

Paper: Although a variety of different terminal devices, socket designs, and other components may be used to construct either type of device, BP and MYO prostheses are viewed as conceptually distinct treatment categories due to their different underlying control modalities. This differentiation between BP and MYO prostheses based on their control modalities might suggest that each design offers relative advantages compared to the other and thus should be considered equally valid options for patient treatment. However, this view is not widely accepted within the United States healthcare system, where prostheses are viewed on a hierarchy of complexity and value [2]. Under this model, insurance policies may prioritize more “basic” technologies (i.e., BP prostheses) in favor of more “advanced” technologies (i.e., MYO prostheses). This tiered categorization implies that BP devices should be considered the standard of care, leading to the implementation of policies in which MYO prostheses are completely excluded from coverage or covered only if BP devices are shown to be insufficient for supporting a patient’s functional goals [3,4,5,6]. These policies may contribute to health disparities based on an individual’s ability to pay for their care, making certain prosthetic technologies unavailable to those who are financially disadvantaged [7].

Comment: Body powered arms are able to successfully address true medical problems as opposed to lack of hand, which is not a strictly medical condition (explanations including study references: [link]). They are far more effective than myoelectric arms when used under real life and work conditions, also because the high control error rates that have not been addressed by research and development [link] and costs incurred by dropped items if the myoelectric arm is consistently used (or suffering and costs incurred by overuse / asymmetry related orthopedic problems if the prosthesis is not consistently used) [link] has not been addressed by manufacturers and research. With that, insurances in my view are absolutely correct to prioritize body powered prostheses where applicable howevermuch the reasons they list may be wrong or poorly argued which appears to make people think they are not correct [link]. That said, a body powered prosthesis has to be built correctly [link], possibly not overly relying on commercial parts, and its prescription has to be correctly explained and argued, in order to achieve adequate acceptance and understanding.

Due diligence is a must, also for insurances. As all medical insurance clients pay into these insurances, it is with representation of the rights of all clients that insurances must ensure a fair process. It is clear that a less insanely priced and usually objective better body-powered prosthesis first has to be ruled out as viable prosthetic provision before one starts spending vast amounts of money to pay for essentially less functional myoelectric arms [link]. That is a logic consideration and one that follows due diligence.

The insurance texts itself do not indicate discrimination of poor people. The details of the cited insurance policies in the study under review here [1] with relation to [3] are that “Myoelectric upper extremity prosthetic devices are considered medically necessary when ALL of the criteria set forth in (A) and (B) below have been met: Selection criteria The individual has an amputation or absence of a portion of an arm; and The individual has sufficient ability to operate the higher level technology effectively; and A standard body-powered prosthetic device cannot be used or is insufficient to meet the functional goals and needs of the individual; and A myoelectric device is likely to help the individual regain or maintain function better than a standard body-powered prosthetic device; and The remaining musculature of the affected arm contains the minimum microvolt threshold to allow operation of a myoelectric device; and (…)”. With relation to reference [4], we read that “Myoelectric upper limb prosthetic components may be MEDICALLY NECESSARY when all of the following conditions are met: The individual has an amputation or missing limb at the wrist or above (eg, forearm, elbow), and Standard body-powered prosthetic devices cannot be used or are insufficient to meet the functional needs of the individual in performing activities of daily living, and  The remaining musculature of the arm(s) contains the minimum microvolt threshold to allow operation of a myoelectric prosthetic device, and The individual has demonstrated sufficient neurological and cognitive function to operate the prosthesis effectively, and The individual is free of comorbidities that could interfere with function of the prosthesis (eg, neuromuscular disease), and Functional evaluation indicates that with training, use of a myoelectric prosthesis is likely to meet the functional needs of the individual (eg, gripping, releasing, holding, coordinating movement of the prosthesis) when performing activities of daily living. This evaluation should consider the individual’s needs for control, durability (maintenance), function (speed, work capability), and usability, and The amputee has been evaluated by an independent qualified professional to determine the most appropriate prosthetic components and control mechanism (eg, body-powered, myoelectric, or combination of body-powered and myoelectric). The independent qualified professional has verified that the amputee meets all the medical necessity criteria for the device.” For reference [6] the text states that “An upper extremity prosthetic for amputations is proven and Medically Necessary when all of the following criteria are met: – Member has a traumatic or surgical amputation of upper extremity or a congenital absence or defect; and – Prosthetic replaces all or part of a missing limb; and – Prosthetic will help the member regain or maintain function; and – Prosthetic device is ordered by or under the direction of a physician; and – Prosthetic needs evaluated for member by a healthcare professional with appropriate prosthetic qualifications and training under the supervision of the ordering physician; and – Member is willing and able to participate in the training for the use of the prosthetic; and – Member with expected rehabilitation potential undergoes functional assessment [including Activities of Daily Living (ADLs) and Instrumental ADLs (IADLs)] evaluation An upper extremity Myoelectric Prosthetic for amputations above the wrist is proven and Medically Necessary in certain circumstances. For medical necessity clinical coverage criteria, refer to the InterQual® CP: Durable Medical Equipment, Prosthetics, Myoelectric, Upper Extremity, Above the Wrist (Custom) – UHG.”, whereas the Interqual can be found to state something along the lines of “Computerized Prosthetic Limbs For the purposes of this policy, the terms computerized, bionic, microprocessor, or myoelectric prostheses are considered the same. Computerized Prosthetic limbs are a covered health care service when all of the following criteria are met: Each of the criteria in the Prosthetic will help the member regain or maintain function; and Devices section are met; and Member is evaluated for his/her individual needs by a healthcare professional with the qualifications and training to make an evaluation under the supervision of the ordering physician; (documentation should accompany the order); and Ordering physician signs the final prosthetic proposal; and The records must document the patient’s current functional capabilities and his/her expected functional rehabilitation potential, including an explanation for the difference, if that is the case. (It is recognized within the functional classification hierarchy that bilateral amputees often cannot be strictly bound by functional level classifications); and Prosthetic replaces all or part of a missing limb; and Prosthetic will help patient regain or maintain function; and Member is willing and able to participate in the training for the use of the prosthetic; (especially important in use of a computerized upper limb); and Member is able to operate the stimulator of the physically function at a level necessary for a computerized prosthetic or microprocessor; and, e.g., hand, leg, or foot Functional assessment (including activities of daily living (ADLs) and Instrumental ADLs (IADLs)) evaluation and expected rehabilitation potential; and Remaining musculature of the arm(s) contains the minimum microvolt threshold to allow operation of a Myoelectric Prosthetic Device (usually 3-5 muscle groups must be activated to use a computerized hand), no external switch; and Ordering physician authorizes the final prosthetic proposal Myoelectric prosthetic components for hand, partial-hand, and artificial digits below the wrist are considered not medically necessary in members who do not meet the criteria above.” And reference [5] states that “Myoelectric upper limb prosthetic components meet the definition of medical necessity when ALL of the following criteria are met: -The member has demonstrated sufficient neurologic and cognitive function to operate the prosthesis effectively; AND -The member has an amputation or missing limb at the wrist or above (e.g. forearm, elbow); AND -The member is free of comorbidities that could interfere with function of the prosthesis (e.g. neuromuscular disease); AND -The remaining musculature of the arm(s) contains the minimum microvolt threshold to allow operation of a myoelectric prosthetic device; AND –Standard body-powered prosthetic devices cannot be used or are insufficient to meet the functional needs of the member in performing activities of daily living (ADLs); AND -Functional evaluation indicates that with training, use of a myoelectric prosthesis is likely to meet the functional needs of the member (eg, gripping, releasing, holding, coordinating movement of the prosthesis) when performing activities of daily living. This evaluation should consider the member’s needs for control, durability (maintenance), function (speed, work capability), and usability. Myoelectric upper limb prosthetic components do not meet the definition of medical necessity if all criteria listed above are not met.”

The claim in the study that proposes that “These policies may contribute to health disparities based on an individual’s ability to pay for their care, making certain prosthetic technologies unavailable to those who are financially disadvantaged” appears wrong in that the insurances not so much seem to want to place a burden on a prosthetic arm user, but they seem to be interested in seeing proof of proper care before paying money into what risks to be a waste. So not wanting to waste money and discriminating against users are distinctly not the same, neither in cause, nor intent. So in other words, insurance policies cited here in no way can be identified to discriminate against users with a rational basis for a need for myoelectric prostheses – much rather, they state that evidence of need is required.

Manufacturers are to name as responsible agents, all around.

First of all, the actual cause for myoelectric prostheses to be unavailable to some financially disadvantaged people is that myoelectric prostheses are far, far too expensive, which is not the fault of any insurance but due to manufacturer pricing. The manufacturers are at fault here, entirely. And the reason why insurances are (and why users should be) very cautious to buy insanely priced myoelectric prostheses is that they lack true function in a range of ways [link biomimetic, link 8 questions].

So secondly, also the lack of true function of myoelectric prostheses in any medical and orthopedic sense, as I see it, however, is entirely the manufacturers’ fault. Which is known and established.

Furthermore, it is known that the relationship of manufacturers and medical device regulation is a big open problem [link CE, link tripleswivel], so we can justify asking whether due diligence and compliance is in place or not with a close look at prosthetic components.

Manufacturers will be to ask why they prefer to build insufficiently functional devices, whose massive price range rightfully causes insurances not to exclude them but to be cautious. But that is all known!

The real question therefore is why on earth the authors of this here cited study [1]  beat around aunt Katie’s house, instead of taking the bull by its horns: why are manufacturers not blamed, why are their practices not named, and why instead are insurances blamed? Why is a cautious way of an insurance to deal with an application of a user for a particular device interpreted as or equated with rejection by the authors? This can only occur when the authors clearly know that a body-powered device is sufficient, or that they lack any evidence, and they still want the other device. Then however it will be up to the user to negotiate a reasonable price with the manufacturer. I had prices drop when I asked for a lower price when ordering and paying myself and not ordering via insurance. When I can do that, anyone can.

Paper: Setting aside any practical constraints imposed by payer restrictions, prosthesis prescription practices would ideally be guided by a determination of what is best for each individual patient based on a wide variety of considerations. However, there is limited empirical evidence available to direct these decisions. This point is demonstrated by two systematic literature reviews which identified 27 experimental studies comparing BP and MYO prostheses published through 2016, of which only 11 used functional or laboratory-based assessments. These reviews failed to substantiate whether BP or MYO prostheses provide a significant general advantage over the other based on this small body of evidence [9, 10]. Similarly, the 2022 VA/DoD Clinical Practice Guideline for the Management of Upper Extremity Amputation Rehabilitation states that there is insufficient evidence to recommend “any specific control strategy, socket design, suspension method, or component” [11] due to low confidence in the quality of existing evidence.

Comment: in 2008, I became a user of a prosthetic arm and as I found out within seconds, not only were the commercial parts woefully weak and under engineered, also these problems were then anything but new, so as it appeared I “inherited” that mess also because generations of previous users had failed to complain properly.

As I found then, no academic or industrial developer even expressed a marginal interest in real work and use aspects of a prosthetic arm, regardless of control technology. To these, we seemed and still seem like mentally unstable asocial freaks, which may be why they support the modern day freakshow “Cybathlon” [link] and why they find it totally adequate to sell junk parts for absolutely insane prices [link].

So with that, I had gone ahead to design, patent and build my own stuff [link].

As a result of that own work, I clearly enjoy massive function and comfort gains of my body-powered prosthesis, while my insurance, given that body powered parts are already considerably cheaper generally albeit crappy, still saves around 60’000 USD per ten years when relying on my own body powered parts. Clearly, I do not need 20-30 cable repairs per year any more but about 1 repair per year, and that can be done by myself using cheap quality bicycle industry parts. So I vastly outperform body powered steel cable setups by Otto Bock, Fillauer, etc., with AliExpress componentry and a bit of hand stitching and cheap 3D plastic printing. The current materials cost for repairs of my prosthetic arm constitutes a fraction of what even just the parking costs me for clinic repairs of my prosthesis when using commercial parts, when added up over a year, so your bet is as good as mine when considering whether to use Otto Bock or Fillauer or such (read about Fillauer and their triple swivel component [link]). My wrist connector is mildly overengineered, so I wear parts there that now are some 15 years old and no problems, no repairs, maybe a cleaning service every 5 or 10 years. With that comes the realization that there is such a thing as planned obsolescence and that ill-fashioned technical design is still practiced, which I now find relatively easy to identify on prosthetic arm parts.

None of the above cited studies or the study authors will in all likelihood be able to describe or assert how to build a truly sustainable, super comfortable body powered control that works also overhead or in close to body reach spaces [link], however, nor will  they be able to build what I call a useful body-powered prosthetic arm. With that, it may not be too interesting to consider what is best for a patient as priority, but the priority as far as I see it should be what can be done to maximize the build quality of prosthetic arms. That is not a listed or visible priority, and to be honest, it is that, which shows.

Paper: While differences in sensory feedback between prosthesis types have been described in the literature through anecdotal evidence [9], no studies have empirically compared feedback available to people with upper limb absence using their own prostheses. To address this gap, we asked participants to complete grasping tasks with their prescribed prosthesis and anatomical limb under different feedback conditions using a custom haptic device [13].

Comment: The comparison of sensory aspects of body-powered versus myoelectric sensory feedback was demonstrated at the Cybathlon 2020 prosthetic arm race. The body-powered 3D-printed plastic arm designed and built by Andrej Dukic won there.

Paper: (…) we asked prosthesis users to complete surveys about perceived prosthesis ownership and agency, as well as a residual limb length estimation task focused on perceptual adaptations to limb absence and prosthesis use [24]. In this paradigm, overestimation of residual limb length while wearing a prosthesis indicates that the user’s perception of their residual limb has expanded into the space occupied by the prosthesis. This overestimation may be viewed as a metric of embodiment [25] (although it is by no means the only option for measuring embodiment [21]). Residual limb length estimation was also performed without the prosthesis to indicate whether this overestimation is retained when the prosthesis is removed. None of these outcomes significantly differed between BP and MYO prosthesis users, as there was significant variability in responses. Even within the three participants who used both BP and MYO prostheses, results were inconsistent (Fig. 2A). This suggests that other participant factors may contribute to embodiment. Here, we found several significant trends that were driven by participant characteristics other than prosthesis type. One influential characteristic was cause of limb absence (acquired vs. congenital). For example, participants with acquired limb loss tended to overestimate their residual limb length both with and without the prosthesis, but participants with congenital limb absence estimated more accurately (Fig. 2B). Additionally, greater residual limb length estimation error when not wearing a prosthesis was correlated with increased hours of daily prosthesis wear. Collectively, our results could support the hypothesis that prosthesis embodiment is also dependent on an individual’s experiences with limb loss and prosthesis use, not just the features of their prosthesis. If this conjecture is validated in future work, it might de-emphasize the need to consider prosthesis design or control strategy as the primary means to promote embodiment.

Comment: Telescoping is nothing new or unknown (e.g., [2]). With an acquired amputation, I may perceive a longer than real residual limb because of vivid phantom sensations. It has nothing to do with embodiment from where I am standing.

Paper: We quantified upper limb and trunk range of motion in BP and MYO prosthesis users during ADLs that required reaching in different planes and manipulating objects with various sizes and shapes [15]. Participants used greater trunk lateral lean during deodorant application when using a BP prosthesis compared to a MYO prosthesis. Additionally, BP users had greater trunk axial rotation and lower shoulder elevation relative to MYO users when placing a box on a high shelf. Otherwise, range of motion did not significantly differ based on prosthesis type. The observed differences may be attributed to reduced shoulder mobility from the harnesses of BP prostheses, which decreases the wearer’s reachable workspace [42, 43] and necessitates increased trunk motion to orient the arm during task performance.

Comment: Trunk motion with a body-powered prosthesis is greatly reduced when one implements the control with a good sound and well engineered build [link]. (Only) then, close to body or overhead reach are easily performed [link]. The technical aspect here is that a figure-9 harness or similar designs will compress soft tissues over significant stretch/extension of the arm, before control of a tensioned cable can be achieved, whereas a stiff shoulder brace massively reduces the extension an arm has to perform before the device is actuated [link]. While it is commonly suggested that trunk motions are somehow bad, the ones that actually matter from a user centered orthopedic view [link] aren’t tasks that are rarely done or done with minimal weight, such as using deodorant under the armpits, which typically may be done once a day.

Paper: Regardless of which outcomes are identified as the most important for a specific user, those metrics might not actually be relevant to healthcare policies in the United States. These policies are informed, at least in part, by the established set of outcomes available for demonstrating value in providing patients with a prosthesis. Since most existing clinical evaluations rely on expert observation or time-based tests, those particular outcomes play a substantial role in determining what a policy will cover. Regrettably, these evaluations fail to capture subtleties in user performance related to movement quality or overall kinematic patterns, as biomechanical outcomes cannot be quantified through visual inspection alone and may only correlate with a limited range of clinical test scores [41, 49, 50]. Current clinical evaluations also do not obviously address more foundational user experiences like embodiment or integration of sensory feedback. If these outcomes are more conclusively shown in future work to be relevant for characterizing individual user performance, it will be critical to include them in clinical assessments as a way to reshape policy priorities.

Comment: since most myoelectric or “bionic” arm users come with very high expectations, and since objectively and for any real work, these are junk [link], it is understandable that real use with high use hours that extends over years is to be assumed to be relatively rare for myoelectric arms. Even fans, aficionados, die hard supporters, are disappointed [link]. Studies are done exclusively with willing participants, whereas anyone else that may have some relevant aspects to report is never represented in any study. So insurance payments for junk arms that we know will be abandoned due to their junk status are problematic, and if even insurances know that, that is where we would expect focused good engineering to start working. As they left the last decades visibly unused for such [link], we can only assume that this is never going to happen. While everyone and their grandmother did and still does their darnedest to talk body powered prostheses bad, to denigrate them to the hilt, this does not change the fact that a physical reality looks, feels, and in fact is, different [link].

Conclusion

The reasons for the ongoing shortcomings in prosthetic arm technology are not technical, they may be financial, but we should really look into dark areas of the souls that are involved as the mechanisms that have reliably failed as over decades are social. There are no rational explanations for wasting millions of Euro for what from a real work delivering user like me looks like device research into useless technology [link]. The way a proper body-powered arm may be built is published, the paper is freely available [link]. We are therefore clearly looking at a purely sociological phenomenon, absent initiative, absent training, absent interest, absent dedication, not being able to listen or read, all sorts of things that have no polite explanations. Any further attempt at explanations that actually label and highlight the true dependencies and responsibilities will have to consider that. Until further notice, as predicted by me 15 years ago, body-powered prosthetic arms are the way to go, if they are well built. If they are not well built then one should complain [link].

For the actual analysis that had the goal of implementing better upper limb prosthetic care we would expect:

  • focus on analyzing compliance of manufacturers with existing medical device legislation, and suggestions for improving medical device laws for better user protection
  • analysis of myoelectric control with a focus on all aspects that prevent control reliability with less than 0,03% control error in real life situations [link] including sweating and affordable myoelectric components that are mechanically robust
  • analysis of why there are no affordable grip covers that are both efficient for deformation and highly resilient with regard to tears and punctures (e.g., [link])
  • analysis of why we lack affordable commercial body-powered components that are systematically correctly designed for full physical load and robust performance, as well as comfort, including shoulder braces and cable mounts
  • analysis of industrial and professional subgroup interests and biases that conflict with the points listed above
  • analysis of insurance needs vice versa the above listed points
  • analysis of conflicts of interest of academic researchers with the above listed points and analysis of how academic research with practically useless focus, methods and results are still promoted and published, and who checks and prevents research that is of no relevance vice versa the above listed points is still approved and financed [link]

Once these are analyzed, a range of recommendations would follow.

[1] S. M. Engdahl, M. A. Gonzalez, C. Lee, and D. H. Gates, “Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses,” Journal of NeuroEngineering and Rehabilitation, vol. 21, iss. 1, p. 138, 2024.
[Bibtex]
@article{engdahl2024perspectives,
  title={Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses},
  author={Engdahl, Susannah M and Gonzalez, Michael A and Lee, Christina and Gates, Deanna H},
  journal={Journal of NeuroEngineering and Rehabilitation},
  volume={21},
  number={1},
  pages={138},
  year={2024},
  publisher={Springer}
}
[2] W. Henderson and G. Smyth, “Phantom limbs,” Journal of Neurology, Neurosurgery, and Psychiatry, vol. 11, iss. 2, p. 88, 1948.
[Bibtex]
@article{henderson1948phantom,
  title={Phantom limbs},
  author={Henderson, WR and Smyth, GE},
  journal={Journal of Neurology, Neurosurgery, and Psychiatry},
  volume={11},
  number={2},
  pages={88},
  year={1948},
  publisher={BMJ Publishing Group}
}

Cite this article:
Wolf Schweitzer: swisswuff.ch - Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses [review]; published 10/01/2025, 06:12; URL: https://www.swisswuff.ch/tech/?p=13817.

BibTeX 1: @MISC{schweitzer_wolf_1738965105, author = {Wolf Schweitzer}, title = {{swisswuff.ch - Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses [review]}}, month = {January}, year = {2025}, url = {https://www.swisswuff.ch/tech/?p=13817}

BibTeX 2: @MISC{schweitzer_wolf_1738965105, author = {Wolf Schweitzer}, title = {{Perspectives on the comparative benefits of body-powered and myoelectric upper limb prostheses [review]}}, howpublished = {Technical Below Elbow Amputee Issues}, month = {January}, year = {2025}, url = {https://www.swisswuff.ch/tech/?p=13817} }